Second International Workshop on Symbolic-Neural Learning (SNL-2018)

July 5-6, 2018
Nagoya Congress Center (Nagoya, Japan)

Uplift Modeling from Separate Labels

Ikko Yamane (The University of Tokyo/RIKEN), Florian Yger (Paris Dauphine University), Jamal Atif (Paris Dauphine University), and Masashi Sugiyama (RIKEN/The University of Tokyo)

Abstract:

Uplift modeling is aimed at estimating the incremental impact of an action on an individual's behavior, which is useful in various application domains such as targeted marketing (advertisement campaigns) and personalized medicine (medical treatments). Conventional methods of uplift modeling require every instance to be jointly equipped with two types of labels: the taken action and its outcome. However, obtaining two labels for each instance at the same time is difficult or expensive in many real-world problems. This poster presents a novel method of uplift modeling that is applicable to a more practical setting where only one type of labels is available for each instance. We show a generalization error bound for the proposed method and demonstrate its effectiveness through experiments.